Occupational Medicine Advance Access originally published online on July 25, 2007
Occupational Medicine 2007 57(7):480-487; doi:10.1093/occmed/kqm058
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Use of prescribed medication at work in employees with chronic illness
1 Department of Human Sciences, Brockington Building, Loughborough University, Loughborough, Leicestershire LE11 3TU, UK
2 Department of Psychology, Goldsmiths College, University of London, New Cross, London SE14 7NW, UK
Correspondence to: Fehmidah Munir, Department of Human Sciences, Brockington Building, Loughborough University, Loughborough, Leicestershire LE11 3TU, UK. Tel: +44 1509 228228; fax: +44 1509 223940; e-mail: f.munir{at}lboro.ac.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionFundingConflicts of interestReferences |
|---|
Background This study examined factors associated with the use of prescribed medication at work.
Methods Questionnaire survey of employees with diagnosed chronic illnesses from four UK organizations. Data were collected on type of chronic illness, health status, health beliefs, work limitations, occupational health support, general practitioner (GP) and line manager support. Data were analysed using univariate logistic regression.
Results A total of 1474 employees with chronic illness participated. Medication use at work (yes versus no) was predicted by age, pain, diagnosis of heart disease, medication use at home, benefit of prescribed medication to health, ease of using medication at work, practical support from families and practical and emotional support from GP and line manager. In a multivariate logistic regression model, medication use at work was predicted by medication use at home and ease of using medication at work only.
Conclusions The ease of taking medication at work was found to be a key predictor of medication use at work, suggesting occupational health may play a vital role in finding ways to support employees in their usage of medication. This may be for example by providing help and guidance in storing medication at work and encouraging employees to disclose medication use to employers and managers where necessary. Occupational health services can help create a workplace culture that places a high value on health, educating staff on the value of looking after their health and the benefits of following advice.
Keywords Chronic illness; medication use; support; workplace intervention
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingConflicts of interestReferences |
|---|
As the incidence of chronic illnesses increases, the number of prescriptions for medication has increased sharply [1–5]. Evidence suggests that use of medication and adherence relies heavily on patients and the health services [6]. Although certain patient characteristics are associated with medication use [7,8], it is now understood that social and psychological variables are among the most significant factors that influence medication use, such as physician–patient communication, illness perceptions, health beliefs and support from general practitioners (GPs), families and significant others [9,10]. However, for many patients, the majority of their day-to-day use and management of medication takes place at work [11]. Most research in this area has not considered the effect of work-related factors on medication use.
It is possible that medication use at work is influenced by work-related social and psychological variables, such as work limitations, line manager support and the opportunity to take prescribed medication at work. Some evidence suggests that for workers managing anxiety and depression, the work environment plays a pivotal role in their medication use and adherence [1–5]. For example, Haslam et al. [1–3] found that side-effects of psychotropic medication for those with anxiety and depression interfered with work performance and non-compliance was reported to be widespread. Lack of information and workplace support was also associated with non-compliance. Haslam et al. [2] argued that lack of compliance can present a serious health and safety risk in the workplace in terms of not only increasing the risk of further health deterioration, but also the ‘knock-on’ effects this may have on sickness absence, presenteeism and work productivity. Further research is needed to explore whether these factors are isolated to medication use for anxiety and depression or extend to other illnesses. Specifically, work-related factors associated with prescribed medication use at work among employees with chronic illness are of importance given the increasing prevalence and enduring nature of chronic illnesses [5]. This study aimed to examine the role of workplace support, GP support, health perceptions of medication use at work and the influence of work performance on medication use at work among employees with chronic illness.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingConflicts of interestReferences |
|---|
Participants were employees from four organizations across three sectors: local government, transport and manufacturing (two companies). To ensure anonymity, employees were randomly sent a questionnaire through their occupational health departments. We approached all employees in the two manufacturing companies (5000 employees) and randomly selected 1:3 employees in the local government (employing 21 000 employees) and 1:2 employees in the transport organization (employing 12 000 employees). Completed questionnaires were returned directly to the research team. To monitor overall response rates, the questionnaire asked all employees independent of their health status for demographic and job-related details. Employees managing a chronic illness were asked additional questions about their health and work.
The questionnaire asked participants to report any medically diagnosed chronic illness currently experienced and to indicate which primary condition (if more than one was listed) most affected their work. This measure is consistent with other self-report measures of chronic illness [11–13]. A total of 17 different groups of chronic illnesses were identified from the sample. Nine illnesses were clearly identified: asthma, irritable bowel syndrome, diabetes, migraine, thyroid disease, inflammatory bowel disease, multiple sclerosis, chronic fatigue syndrome, epilepsy, HIV and hepatitis (B and C). For depression and anxiety, participants were grouped if they reported either depression, anxiety or a combination of both. The International Classification of Diseases [14] was used to classify the following illnesses: musculoskeletal pain consisted of participants reporting pain anywhere along the musculoskeletal system (e.g. back, shoulders, neck, arms, elbows, wrist and lower limbs). Heart disease included myocardial infarction, angina, heart failure, stroke and hypertension. For cancer, participants were included if they reported any type of cancer. Eye problems included glaucoma, sensory neuropathy and blepharitis. Reproductive and gynaecological problems included endometriosis and menstruation problems. For arthritis and rheumatism, participants were included if they reported any form of arthritis, spondylitis and fibromyalgia. A further group was classified as ‘other’, which represented smaller numbers of reported other chronic illness. Employees were also asked how long they had their illness (years).
Prescribed medication use at home and at work was assessed by a modified version of the illness symptoms self-management behaviours scale [15,16]. Participants were asked to rate how closely they were following their doctor's advice in taking prescribed medication related to their chronic illness at home and at work (two items). Responses were measured on a 10-point Likert scale (‘not closely’ to ‘very closely’) and had an internal consistency of 
= 0.84. Based on the same scale [15], participants were also asked to rate how beneficial they felt it was to their health to take medication at work (one item, measured on a 10-point Likert scale from ‘not beneficial at all’ to ‘very beneficial’). They were also asked if it was easy for them to take medication at work (yes or no).
Health status was assessed by a modified version of the pain severity scale (two items, = 0.66); fatigue (one item) and health distress (four items, = 0.89). All items were taken from the Medical Outcomes Study survey of health status [17] and measured on a five-point Likert scale. A mean score was calculated for the health distress scale. Severity of symptoms (one item) was also measured (mild to severe).
The Work Limitations Questionnaire [18] was used to assess the degree to which employees' symptoms of chronic illness interfered with specific aspects of job performance. The questionnaire asks employees to rate on a five-point Likert scale, their level of difficulty (or ability) to perform 25 specific job demands corresponding to four scales: time management (five items, = 0.88), physical demands (six items, = 0.90); mental interpersonal (nine items, = 0.92) and output demands (five items, = 0.94). A total scale score was calculated to indicate overall work limitation, where a higher score indicated more work limitation [18].
Support consisted of two forms of workplace support: practical (giving information, help and advice) and emotional support (sympathy and understanding). These were measured with four items each, representing support received from line manager, occupational health, family and GP in the management of chronic illness. Items were measured on a five-point Likert scale (‘no support’ to ‘a great deal of support’) and had an internal consistency of = 0.83 (line manager) = 0.81 (occupational health), = 0.88 (family) and = 0.87 (GP).
Data were collected on age (years), sex, tenure (length of employment, years), employment status (part/full-time) and education (none, GCSE, AS level A level or equivalent, degree, higher degree).
Correlations were computed for all variables. Those which significantly correlated with the medication use at work were subjected to logistic regression analysis to examine their association with medication use at work (as the distribution of responses showed high relative frequencies of minimum and maximum scores indicating that participants were either likely to take medication or not, scores 1–5 were dichotomized into ‘no’ and scores 6–10 were dichotomized into ‘yes’). For analyses, age was classified as 18–30, 31–43, 44–56 and 57–69. A score of 1 was given if a chronic illness was present and a score of 0 if any other illness was present. Those proven to be significant predictors at P < 0.001 (Bonferroni correction) were entered into a multiple logistic regression analysis to identify the strongest predictors.
A review of each organisation's policy in medication use showed the transport organization required employees to disclose medication use for health and safety reasons. Therefore, to control for organizational differences in medication policies, each organization was entered as a co-variate in step 1 of the analyses using dummy coding (1 = organization with medication policy, 0 = all other organizations).
Ethical approval was granted by The Institute of Work, Health & Organisations, University of Nottingham, local ethics committee.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingConflicts of interestReferences |
|---|
A total of 1474 participants reported at least one chronic illness giving a response rate of 28% for completed returned questionnaires. The remaining sample did not report a chronic illness and were excluded from further analysis.
Table 1 reports demographic and illness-related variables. This was compared with data obtained from each organization's human resources department (non-responders). Participants with chronic illnesses did not significantly differ from their respective colleagues in terms of gender and occupational status (all P > 0.05). However, those reporting heart disease and arthritis and rheumatism were significantly older than non-responders (P < 0.05). Across organizations, musculoskeletal pain was the most reported condition. For administration and manufacturing A organizations, arthritis and rheumatism were the second most prevalent condition. For transport and manufacturing B, asthma was the second most reported condition. Across the sample, 85% (n = 1256) reported using prescribed medication at work related to their primary chronic illness affecting work. Table 2 reports the means and standard deviations for medication use at home and at work and were compared between the chronic illness groups using multivariate analysis of co-variance (MANCOVA). Illness group was entered as the independent variable. Age, gender, organization, education, severity of illness and length of time managing an illness were entered as co-variates. A Bonferroni corrected alpha level of P 
0.001 was accepted as statistically significant. MANCOVA revealed a significant main effect between groups [F(1, 16) = 4.45; P < 0.001]. Tukey's post hoc analyses showed those with musculoskeletal pain were less likely to use medication both at work and at home than all other groups (P < 0.001). No other significant group differences were found. There was a significant main effect within group [F(1, 1112) = 163.14; P < 0.001], indicating that medication behaviour significantly differed between home and work. T-tests revealed those with arthritis, asthma, musculoskeletal pain, depression and anxiety, diabetes, migraine and irritable bowel syndrome were more likely to take medication at home than at work (all P values <0.001).
|
|
Table 3 shows the correlations between the outcome variable ‘medication use at work’ and all other variables. The correlations range from r = 0.01 to r = 0.60, indicating no highly significant problems with multicollinearity. The adjusted odds ratios of predictors for medication use at work are presented in Table 4. For demographic and illness-related variables, only age and pain were related to medication use at work (P < 0.001). Participants with diagnosed heart disease were more likely to use medication at work and those with diagnosed musculoskeletal pain and depression and anxiety were less likely to use medication at work compared with other chronic illnesses. The use of medication at home increased the likelihood of medication use at work and the easy use of medication at work was also predictive of medication use at work. The perceived benefit of prescribed medication to health was also associated with medication use at work. In terms of support, instrumental support from line manager, GP and from family were associated with medication use at work. Support from occupational health was not associated with medication use at work. Work limitation was also not associated with medication use at work. The variables associated with medication use at work were then entered into a multiple logistic regression model. A Bonferroni corrected alpha level of P
0.001 was accepted as statistically significant.
|
|
Table 5 shows the multivariate model of significant predictors (
2 = 18.24, df = 2; P < 0.01). The final model (sensitivity 63% and specificity 95%) accurately classified 89% of the cases. In the final model, only medication use at home and ease of taking medication at work predicted medication use at work.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingConflicts of interestReferences |
|---|
Our study found that medication use at work was predicted by medication use at home and the ease of using medication at work. About 85% of respondents with a chronic illness reported using medication at work. This high level of usage, combined with the increasing prevalence of chronic illness in the workplace as we face an ageing workforce [19], highlights the need to better understand medication use at work.
Medication use by illness group showed that employees with heart disease were more likely to use medication at work, and employees with arthritis, asthma, depression and anxiety, diabetes, migraine and irritable bowel syndrome were more likely to take medication at home than at work (Tables 2 and 4). In particular, those with musculoskeletal pain and depression and anxiety were less likely to use medication at work compared with all other groups (Table 4). Certain factors associated with musculoskeletal pain, depression and anxiety but not measured in this study, may help to explain the low use of medication by these groups. For example, the stigma associated with depression and anxiety or the possible side-effects of psychotropic medication use on work performance may deter employees from taking their medication [1–3]. However, it seems more probable that for these illnesses medication may be required once or twice a day and therefore outside of working hours.
Our findings on predictors of medication use are in line with other studies on medication use and adherence [7,8], revealing participants' age, severity of pain and the perception that prescribed medication use is beneficial to health independently predicted medication use at work. In our study, those with mild pain were more likely to use medication than those with severe pain, perhaps indicating those using medication for pain control are therefore more likely to report lower pain than those who are not. This highlights the importance of reinforcing the benefits of medication use for those who are prescribed medication. While this may initially fall within the remit of a GP prescribing the medication, primary care providers often have limited contact time with the individual, which is often not sufficient for attitude change should an individual be sceptical or place a low value on the beneficial properties of medication. Organizations on the other hand could play a vital role in this behaviour changing process, through ensuring there is a culture that places a high value on health.
In addition, both emotional and practical support from participants' GP and line manager independently predicted medication use at work. Our results are consistent with previous research in that support (practical and emotional) strongly relate to medication use and other self-managing behaviours [9–11], and extends those findings by indicating the importance of support in the use of medication at work.
Our final multiple model suggests that using medication at home and the ease of using medication at work were the most parsimonious predictors of using medication at work. Although GP support was the strongest independent predictor of support relating to medication use at work, it was not retained in our final model, suggesting its' influence was absorbed by using medication at home. Although employees were more likely to use medication at work if the work environment facilitated the use of medication, i.e. made it easy for employees to take medication, support from occupational health did not predict medication use at work. However, it is not known from the current study, whether occupational health services made the facilitation of medication use easier for employees. For example, implementing a secure box or fridge for medications or providing a private room for employees to take medication or by creating a culture of workplace health that encourages employees to manage their illness. Many of these activities may be undertaken by and implemented by both management and occupational health services.
A review of each organizations' policy on medication use at work revealed only one organization had a policy which required employees to disclose medication use for health and safety reasons. However, as the influence of organizational policies was controlled for in the analyses, further research is required to understand organizational practices for medication use at work, particularly to what extent certain policies and practices encourage or deter employees in taking their medication at work, and whether lack of medication use have negative physical or psychological health outcomes such as pain, fatigue and stress. Further longitudinal intervention-based studies are needed to delineate this relationship.
While we achieved a below average response rate for mailed surveys of this type [20,21], discussions with organizational stakeholders indicate that response rates for questionnaires outside of annual employee surveys are in the region of 27–31% due to survey fatigue [22]. Survey fatigue is an increasing problem faced by researchers conducting organizational-based research, despite usage of response-inducing techniques [22]. As the study relied entirely on self-report data in identifying those with chronic illnesses, this may have resulted in under-reporting of chronic illnesses, leading to a somewhat lower response rate or an under-representation of those not adhering to medication use. It is also not possible to know whether employees felt uncomfortable in completing a questionnaire about their illness at work or felt their illness posed no problem at work or simply chose not to fill it in due to lack of time. Nevertheless, demographic comparisons between responders and non-responders indicated no serious problems with response bias.
Further longitudinal research using clinical populations is required to understand what factors encourage or deter prescribed medication use at work. Such studies can capture information on medication use compliance, changes in medication use, employment status and support and the fluctuations of physical, social and psychological correlates. Such information can help occupational health professionals design, implement and evaluate appropriate intervention strategies in medication use and management to help promote such employees' well-being and minimize subsequent sickness absence. The focus of this study is worthy of further attention not only from a research perspective but also from the point of collaboration between health care professionals and industries.
Implications for clinical practice: the ease of taking medication at work was found to be a key predictor of medication use at work, suggesting occupational health may play a vital role in finding ways to support employees in their usage of medication. For example, by providing help and guidance in storing medication at work and encouraging employees to disclose medication use to employers and managers where necessary. Occupational health staff also play an important role in working with employees and their line managers and team to manage possible side-effects of medication and their possible impact on work performance and safety. The workplace is an ideal environment in which to influence and educate individuals on the value of medication and the benefits illness self-management can accrue. Occupational health services can help create a workplace culture that places a high value on health, educating staff on the value of looking after their health and the benefits of following advice given and work closely with GPs to increase medication compliance. With greater opportunity of contact with individuals, and a number of mediums through which to shape behaviour change (e.g. posters, emails, one-to-ones), occupational health professionals are well positioned to enhance medication use at work.
Key points
|
|
Funding |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingConflicts of interestReferences |
|---|
European Social Fund (EO4).
|
Conflicts of interest |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingConflicts of interestReferences |
|---|
None declared.
|
Acknowledgements |
|---|
We thank Amanda Griffiths, Sara Cox and Stavroula Leka for their suggestions at the initial design of the study. We also thank participating organizations for access.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingConflicts of interestReferences |
|---|
- Haslam C, Atkinson S, Brown S, Haslam R. Anxiety and depression in the workplace: effects on the individual and organisation (a focus group investigation). J Affect Disord (2005) 88:209–215.[CrossRef][Web of Science][Medline]
- Haslam C, Atkinson S, Brown S, Haslam R. Perceptions of the impact of depression and anxiety and the medication for these conditions on safety in the workplace. Occup Environ Med (2005) 62:538–545.
[Abstract/Free Full Text] - Haslam C, Brown S, Atkinson S, Haslam R. Patients' experiences of medication for anxiety and depression: effects on working life. Fam Pract (2005) 21:204–212.[CrossRef][Web of Science]
- Potter WZ. Psychotropic medicines and work performance. J Occup Med (1990) 32:355–361.[CrossRef][Web of Science][Medline]
- World Health Organization. Health Promotion and Chronic Diseases: Discovering a New Quality of Health (1992) Copenhagen: World Health Organization Regional Office for Europe.
- Korff MV, Glasgow RE, Sharpe M. Organising care for chronic illness. Br Med J (2002) 325:92–94.
[Free Full Text] - Devine JW, Farley JF, Hadsall RS. Patterns and predictors of prescription medication use in the management of headache: findings from the 2000 medical expenditure panel survey. Headache (2005) 45:1171–1180.[CrossRef][Web of Science][Medline]
- Balkrishnan R. The importance of medication adherence in improving chronic-disease related outcomes—what we know and what we need to further know. Med Care (2005) 43:517–520.[CrossRef][Web of Science][Medline]
- Bryne M, Walsh J, Murphy AW. Secondary prevention of coronary heart disease: patient beliefs and health-related behaviour. J Psychosom Res (2005) 58:403–415.[CrossRef][Web of Science][Medline]
- Gonzalez JS, Penedo FJ, Antoni MH, et al. Social support, positive states of mind, and HIV treatment adherence in men and women living with HIV/AIDS. Health Psychol (2004) 23:413–418.[CrossRef][Web of Science][Medline]
- Munir F, Leka S, Griffiths A. Dealing with self-management of chronic illness at work: predictors for self-disclosure. Soc Sci Med (2005) 60:1397–1407.[CrossRef][Web of Science][Medline]
- Dewa CS, Lin E. Chronic physical illness, psychiatric disorder and disability in the workplace. Soc Sci Med (2000) 51:41–50.[CrossRef][Web of Science][Medline]
- Lerner D, Amick BC, Malspeis S, Rogers WH. A national survey of health-related work limitations among employed persons in the United States. Disabil Rehabil (2000) 22:225–232.[CrossRef][Web of Science][Medline]
- World Health Organization. International Classification of Diseases—Version 10 (1999) Geneva: World Health Organization.
- Clark NM, Dodge JA. Exploring self-efficacy as a predictor of disease management. Health Educ Behav (1999) 26:72–89.
[Abstract/Free Full Text] - Lorig K, Stewart A, Ritter P, Gonzalez V, Laurent D, Lynch J. Outcome measures for Health Education and Other Health Care Interventions (1996) Thousand Oaks, CA: Sage.
- Stewart AL, Hays RD, Ware JE. Health perceptions, energy/fatigue, and health distress nmeasures. In: Measuring Functioning and Well-Being: the Medical Outcomes Study Approach—Stewart AL, Ware JE, eds. (1992) Durham, NC: Duke University Press. 143–172.
- Lerner DJ, Amick BC III, Rogers WH, Malspeis S, Bungay K. The work limitations questionnaire: A self-administered instrument for assessing on-the-job work disability. Med Care (2001) 39:72–85.[CrossRef][Web of Science][Medline]
- Ilmarinen JE. Aging workers. Occup Environ Med (2001) 58:546–552.
[Free Full Text] - Baruch Y. Response rate in academic studies—a comparative analysis. Hum Relat (1999) 52:345–353.
- Roth PL, BeVier CA. Response rates in HRM/OB survey research: norms and correlates, 1990–1994. J Manage (1998) 24:97–117.[CrossRef]
- Ibeh K, Brock JK-U, Zhou YJ. The drop and collect survey among industrial populations: theory and empirical evidence. Ind Market Manage (2004) 33:155–165.[CrossRef]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||